A Model for Cross-Domain Opinion Target Extraction in Sentiment Analysis

Opinion target extraction is one of the core tasks in sentiment analysis on text data. In recent years, dependency parser–based approaches have been commonly studied for opinion extraction. However, parsers are limited by language and grammatical constraints. Therefore, this work, a sequential pattern-based rule mining model, which does not such constraints, proposed cross-domain from product reviews unknown domains. Thus, knowing domain while extracting targets becomes no longer requirement. The model also reveals difference between concepts aspect, confused literature. consists two stages. first stage, aspects extracted using rules automatically generated source transferred domains to domain. Moreover, aspect pruning applied further improve performance second among at former stage was evaluated several benchmark datasets different compared against experimental results revealed that can be with higher accuracy than those previous works.